The effect of significant other involvement in treatment for substance use disorders: A meta-analysis

Abstract

Objective: Substantial research has accrued in support of a key role for social processes in substance use disorders (SUD). Researchers have developed a range of interventions that capitalize on these social processes to bolster treatment outcomes by involving significant others (e.g., romantic partners, family, friends) in SUD treatment. Yet dissemination of these treatments to many contexts has been slow, and information on their broad efficacy is lacking. This meta-analysis aims to quantify the effect of significant other involvement in SUD treatments above and beyond individually-based therapies.

Method: A total of 4,901 records were screened for randomized controlled trials examining the effect of Significant Other Involved SUD Treatments (SOIT) versus individually-based active comparator treatments. Our search yielded 77 effect sizes based on data from 2,115 individuals enrolled in 16 independent trials.

Results: Findings indicated a significant effect of SOIT above and beyond individually-based active comparator treatments for reducing substance use and substance-related problems, d = 0.242, 95% CI [0.148, 0.336], I² = 10.596, Q(15) = 16.778. This effect was consistent across SOIT treatment types and endured 12-18 months after the end of treatment. Analyses of raw mean differences indicated that this effect translates to a 5.7% reduction in substance use frequency-the equivalent of approximately 3 fewer weeks a year of drinking/drug use.

Conclusion: Findings indicate a significant advantage for SOIT in SUD treatment, and hold interesting conceptual implications for theories of SUD maintenance. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Figure 1.

Number of studies referencing significant-other involved SUD interventions displayed according to year of publication. Note. Database search terms used to produce the above results are outlined within the “search strategy” portion of the methods section.

Figure 2.

PRISMA flow diagram illustrating the process of identifying eligible studies. Note. Studies meeting “other exclusion criteria”: Fourteen studies did not have an experimental and/or comparison treatment group that had at least one clear face-to-face therapy session component. Four studies included treatment modalities where significant-others sought treatment with the aim of engaging and motivating the diagnosed individual to start treatment. Five studies recruited substance using populations that presented with co-morbid severe mental illnesses. Thirteen studies listed Fals-Stewart as a lead author on a study or a grant receiver on the project.

Figure 3.

Funnel plot of studies included in the meta-analysis.

Figure 4.

Effect size table with forest plot. Note. Cohen’s d values reported above represent the average of all effects that could be derived from the sample, averaged across all time points (post-treatment and follow-up) and substance use outcomes (frequency, problems, abstience, quantity, and “other use measure”). Effects subdivided by time point and outcome are presented in the main text.